misc/124342: sbin/ggate/ggated/ggated.c fails to compile under -CURRENT with default warning settings and custom CFLAGs

[Garrett Cooper]

>Number:         124342
>Category:       misc
>Synopsis:       sbin/ggate/ggated/ggated.c fails to compile under -CURRENT with default warning settings and custom CFLAGs
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Jun 06 06:10:04 UTC 2008
>Closed-Date:
>Last-Modified:
>Originator:     Garrett Cooper
>Release:        8-CURRENT
>Organization:
n/a
>Environment:
FreeBSD optimus 8.0-CURRENT FreeBSD 8.0-CURRENT #0: Mon May 19 02:21:07 PDT 2008     root at optimus:/usr/obj/usr/src/sys/OPTIMUS  i386
>Description:
ggated fails to compile with the following error:

/devel/ncvs/src/sbin/ggate/ggated/ggated.c: In function 'disk_thread':
/devel/ncvs/src/sbin/ggate/ggated/ggated.c:759: warning: no return statement in function returning non-void
/devel/ncvs/src/sbin/ggate/ggated/ggated.c: In function 'send_thread':
/devel/ncvs/src/sbin/ggate/ggated/ggated.c:813: warning: no return statement in function returning non-void

[gcooper at optimus ~]$ grep -E 'CFLAGS|CPUTYPE' /etc/make.conf
CFLAGS=-O3 -pipe -fno-strict-aliasing
CPUTYPE=prescott
>How-To-Repeat:
1. Checkout srcs from cvs.
2. cd .../src/sbin/ggate/ggated && make all 
>Fix:
A proposed patch is attached (simple, I know), and I've attached some comments for the author / maintainer of ggate.

I'd like to make sure that a return value of (void*) NULL is appropriate though, in both cases where return values were required...

Patch attached with submission follows:

Index: sbin/ggate/ggated/ggated.c
===================================================================
RCS file: /home/ncvs/src/sbin/ggate/ggated/ggated.c,v
retrieving revision 1.9
diff -r1.9 ggated.c
105a106,108
> /*
>  * XXX: See comment in connection_launch(..).
>  */
106a110,112
> /*
>  * XXX: See comment in bottom of function. 
>  */
579a586,593
> 
> 	/**
> 	 * XXX: Does disk thread really need a return value if it's just
> 	 * tossed here (or is this future work to make into a pthread_create(3)
> 	 * call?)
> 	 *
> 	 * (gcooper)
> 	 */
580a595
> 
682a698,700
> /*
>  * XXX: See comment in connection_launch(..)
>  */
758a777,779
> 
> 	return (void*) NULL;
> 
812a834,843
> 
> 	/**
> 	 * XXX: Is this a correct return value?
> 	 * According to the prototype and the
> 	 * definition above this _must_ have a
> 	 * return value...
> 	 * (gcooper)
> 	 **/
> 	return (void*) NULL;
> 


>Release-Note:
>Audit-Trail:
>Unformatted: